DETECTION OF COAL COMBUSTION PRODUCTS IN STREAM SEDIMENTS BY CHEMICAL ANALYSIS AND MAGNETIC SUSCEPTIBILITY MEASUREMENTS S. Frančišković-Bilinski

Institut “Ruđer Bošković”, Zavod za istraživanje mora i okoliša, Laboratorij za fizičku kemiju tragova OKOLIŠ

1. Introduction The aim of this preliminary study was to apply a rapid and inexpensive, low-field magnetic susceptibility method (MS) to stream sediments, as described by Scholger (1998) and Petrovsky et al. (2000), and to delineate ZAGREB polluted areas in the Kupa river basin. Geochemical characterization of the <63 μm sediment fraction has already  Sava River been carried out (Frančišković-Bilinski, 2007). Increased MS was observed in: (1) The lower stretch of the Mrežnica and Korana rivers, where several elements (U, Sb, Sn, Zr, Nb, S, Na, Ni, Se, Sr, Y, Nb) showed

 Kupa River anomalously high concentrations. This region is located on the Dinaric carbonate platform and the anomalies are of  Kupa River KARLOVAC SISAK anthropogenic origin. (2) At the middle flow of Glina River, where several other elements showed anomalously high concentrations (Fe, Sc, V, Zr, Na, Cu, Ga, Y). The anomalies in this region are of natural origin, influenced by  Korana River  Mrežnica River Supradinaric belt with ophiolites. The MS did not detect extreme Ba anomalies, described in the same drainage basin (Frančišković-Bilinski, 2006). In the present work we concentrated our research on the MS anomaly observed in the lower parts of the Mrežnica and Korana rivers. The study area, including 22 sampling locations, is presented in Figure 1. Sampling station details are listed in Table 1. The pollution source in the Mrežnica river was a large textile factory in Duga Resa (near Karlovac), which burned coal for ~110 y, until 1994; all coal slag and ash were deposited directly into the Mrežnica river. Figure 1. Study area with sampling locations presented by circles (left) and location map for Kupa River drainage basin within Croatia, Slovenia and Bosnia and Herzegovina 2. Materials and methods Stream sediments were collected in April 2007 and the <2 mm fraction was prepared for analysis by air drying, dry sieving and pulverizing. The MS was measured using MS2 (Bartington Instruments, England) and sensor type MS2B. A Perkin Elmer SCIEX ELAN 6100 ICP-MS spectrometer (ACTLABS, Canada) was used to analyse 54 elements with the program Ultratrace 2. Analyses for C, TOC and S were performed using a LECO CS-300 (USA) instrument. Tot. combustion was at 1200ºC, adding W granulate.

Mrežnica River before pollution source Pollution source: Textile factory in Duga Resa 3. Results Increased values of low-field MS (MS, ×10–8 m3/kg) were observed in Mrežnica River, downstream of the pollution source in Duga Resa; in the Korana River, downstream of Mrežnica River inflow; and in Kupa River, downstream of Korana River inflow; and in Pokupsko, ~50 km downstream of the pollution source. MS data are presented in Table 1. The ICP-MS and MS data were divided into four subsets for statistical evaluation: The subset BG consists of only one sampling station (212) and represents the background values for Mrežnica River above the pollution source. Subset A consists of three sampling stations (210, 211, 221) in Mrežnica River, downstream of the pollution source. The subset K consists of eight sampling stations (200, 201, 202, 203, 204, 218, 217, 209) in Kupica River and in Kupa River above the Korana River inflow. Subset DS-K consists of 10 sampling stations (220, 219, 216, 215, 214, 213, 208, 207, 205, 206) in the Korana and Kupa Rivers, downstream of Korana River inflow. In Table 2, statistical parameters for MS and selected elements are presented. The geoaccumulation index, Igeo (=log2(Cn/1.5Bn) where Cn = measured elemental concentration and Bn = background concentration; Müller, 1979) was calculated for subset A. Igeo describes the intensity of contamination of sediments, with respect to metal pollutants (Förstner et al., 1993). The largest Igeo values were observed for B, U, Mo and Zr. Cluster analysis of R-modality was performed on the total dataset to find the relationship between the 54 elements and MS (Fig. 2). MS clustered with B, Mo, Na and U. Significant correlations (>0.90) existed between MS and the following elements: B (0.96); U (0.95); Zr (0.94); Sr (0.93); Na (0.92); Mo (0.92); Ni (0.90). Acknowledgements This research was performed within project No. 4. Discussion 098-0982934-2720 (P.I. I. Pižeta) financed by the Ministry of Science, Education and Sport of According to the Igeo classification of Förstner et al. the Republic of Croatia and bilateral project (1993), in the anomalous region of the present study, Croatia-Austria 2006-2007 (P.I. S. Frančišković- Bilinski and G. Rantitsch). Sampling was sediments are strongly contaminated with U, B, Mo performed within the bilateral project Croatia- and Zr, moderately to strongly contaminated with Al, India 2006-2009 (P.I. H. Bilinski and S.K. Ni and Cu, moderately contaminated with Hg, Na, V, Sarkar) with four participants on the field trip (S. Frančišković-Bilinski, H. Bilinski, S.K. Sarkar Cr and Fe, and uncontaminated to moderately and A. Bhattacharya). Special thanks to R. contaminated with Sr. Scholger and M. Hanesch for their permission to perform MS measurements. G. Rantitsch is The study area is a model area to study the thanked for permission to perform TOC and S behaviour of coal slag and coal ash deposited in a measurements, for use of the statistics program clean karstic tufa-forming river. The MS data illustrate and his helpful critical suggestions. clearly that coal combustion products have been Figure 2. Dendogram obtained by cluster analysis of R-modality for 55 var. transported far downstream from their source. The poor correlation of MS with Fe (0.36) indicates that 5. Conclusions Fe in this river is in neither paramagnetic nor A quick and inexpensive, low-field magnetic susceptibility method (MS) provided an ferromagnetic form. This result suggests that just as indicator of contamination of stream sediments by coal combustion products. Low in soils (Kapička et al., 2001) the properties of the correlation of MS with Fe (0.36) and the absence of magnetite and maghemite in coal-combustion products have altered following sediments suggest that elements other than Fe contribute to magnetic properties. exposure to the river water. Neither ourselves or Cluster analysis of R-modality performed on the total dataset shows that MS data are Kapička et al. (2001) detected magnetite or related to B, Mo, U and Na. A detailed future study of chemical reactions and redox maghemite by XRD yet. The significant correlation of conditions in stream sediments contaminated with coal combustion products is MS with B, Na and Ni is most likely due to formation anticipated. of sodium borate glass during the combustion References process; Ni is known to partition into sodium borate Fermi, 2008. Magnetic susceptibility of the elements and inorganic compounds. Fermi National Accelerator Laboratory, document 4-135, available online: http://www-d0.fnal.gov/hardware/cal/lvps_info/engineering/elementmagn.pdf glass (Kashif et al., 1991). Similarly Mo and U can be Frančišković-Bilinski, S., 2006. Barium anomaly in Kupa River drainage basin. Journal of Geochemical Exploration 88(1-3), 106-109. Frančišković-Bilinski, S., 2007. An assessment of multielemental composition in stream sediments of Kupa River drainage basin, Croatia for evaluating sediment quality incorporated in sodium borate glasses. Several Mo guidelines. Fresenius Environmental Bulletin 5, 561-575. Förstner, U., Ahlf, W., Calmano, W., 1993. Sediment quality objectives and criteria development in Germany. Water Science and Technology 28, 307. and U compounds have positive (uranium (VI) oxide Kapička, A., Jordanova, N., Petrovsky, E., Ustjak, S., 2001. Effect of different soil conditions on magnetic parameters of power-plant fly ashes. Journal of Applied Geophysics 48, 93-102. and molybdenum (VI) oxide) and high positive (Mo Kashif, I., Farouk, H., Aly, S.A., Moustaffa, F.A., Sanad, A.M., Abo-Zeid, Y.M., 1991. Structure and magnetic susceptibility of irradiated sodium borate glasses containing nickel IV, V compounds and U IV, III compounds) MS oxide. Journal of Materials Science: Naterials in Electronics 2(4), 216-219. Müller, G., 1979. Schwermetalle in den Sedimenten des Rheines – Veränderungen seit 1971. Umschau 79, 778-785. (Fermi, 2008) which again could explain the Petrovsky, E., Kapička, A., Jordanova, N., Knab, M., Hoffmann, V., 2000. Low-field magnetic susceptibility: a proxy method of estimating increased pollution of different environmental systems. Environmental Geology 39, 312-318. clustering of MS with Mo and U. Scholger, R., 1998. Heavy metal pollution monitoring by magnetic susceptibility measurements applied to sediments of the river Mur (Styria, Austria). European Journal of Environmental and Engineering Geophysics 3, 25-37.